Radio frequency receivers can be configured to operate within a number of different radio frequency bands. For example a receiver for a mobile station (or cellular telephony device) can be configured to operate within any of the following bands: Global System for Mobile Communications (GSM), 850, 900, 1800, and/or 1900, Wideband Code Division Multiple Access (WCDMA), High Speed Packet Access (HSPA) and/or Long Term Evolution (LTE) Bands 1, 2, 3, etc. This allows a mobile station containing such a receiver to be used in different areas where varying subsets of the above radio frequency hands are supported (e.g. to enable roaming).
Receivers typically incorporate one or more Radio-Frequency Integrated Circuits (RFICs) including a Low Noise Amplifier (LNA) as the first amplifying stage in the receiver. For example, one or more LNAs are typically used to amplify the radio frequency signals gathered by an antenna, and the amplified signals generated by the LNA(s) are then used by other components in the receiver.
Receivers typically include one or more radio frequency (RF) filters located between the antenna and the LNA(s) that form the first amplifying stage of the receiver. FIG. 1 illustrates an exemplary receiver comprising an RF module 100 and antenna 130. RF module 100 comprises an RF Front End Module 132 which in turn includes one or more (up to a total of n) RF filters 110-112 that filter radio frequency signals gathered by antenna 130. RF module 100 also comprises an RFIC 134 which in turn comprises one or more (up to a total of m) LNAs 120-122 that amplify the filtered signals generated by the RF filters 110-112.
As is known from Friis' formula for noise factor, the LNA that forms the first amplifying stage of a receiver dominates the noise figure of the receiver. The LNA that forms the first stage also has a key role in determining the input impedance of the receiver. The input impedance of this LNA must be carefully matched to a certain impedance, as otherwise the performance of an RF filter (e.g. 110-112) preceding the LNA will be degraded. Additionally, an RF filter preceding the LNA will typically have a fixed frequency range which requires the inputs of the LNA to also be matched to that frequency range.
Typically, differential inputs are used in order to get better rejection against interfering signals. Single-ended blocker signals from the antenna port are attenuated by the RF-filters. In addition, in Frequency Division Duplex (FDD) systems, transmitter (TX) leakage from the TX port is attenuated by the duplex filter. Single-ended interfering signals at the antenna and transmitter ports tend to have high common-mode characteristics at the differential LNA input. However, only differential attenuation of the duplex and RF-filter is typically guaranteed according to the specification. These high-level common-mode blocker signals can desensitize the receiver front-end and create undesirable intermodulation products. Since the unmatched LNA input typically experiences up to 6 dB higher blocker signals compared to a matched case if the input impedance is relatively higher at the blocker frequency than at the wanted signal frequency, it is desirable to create a differential LNA topology with good common-mode matching. In addition, if the input matching is wide enough in order to give good matching for far away blocker signals such as TX, Wireless Local Area Network (WLAN) and Bluetooth™, the blocker signals are rejected already at the input. In addition, as mentioned above, the duplex filler performance is degraded in those frequencies if the input is not matched.
As a result, depending on the LNA structure, it may be necessary to utilize matching components external to the RFIC containing the LNA to appropriately set the input impedance and frequency range matching. However, these external matching components can be expensive, and in some cases it is preferable to use an LNA with internal matching capabilities to appropriately set its input impedance and frequency range matching.
Another measure of receiver performance is its sensitivity (reference sensitivity level), which measures the minimum detectable signal level at the receiver input. The signal quality of the received signal is typically determined by bit error rate or throughput. The sensitivity level S is determined by the equation:S=−174 dBm/Hz+10 log(BW)+SNRmin+NR  (1)where −174 dBm/Hz is the available noise power density from an input source at a temperature of 290 K, BW is the channel bandwidth, SNRmin is the required signal-to-noise ratio, and NF is the receiver noise figure. The SNRmin depends on the targeted bit error rate and the modulation method used, for example.
The RF filter preceding the LNA that forms the first amplifying stage in a receiver may have significant insertion loss in some of the radio frequency bands within which the receiver is configured to operate. The insertion loss can cause the receiver to be less sensitive and have a higher noise figure for these radio frequency bands. Since the receiver sensitivity in these radio frequency bands is worse, the range between the transmitter and the receiver over which the receiver may be required to operate is reduced, thus making the cellular network design more challenging and more expensive. In addition, the size of the antenna connected to the receiver may be limited due to space constraints in devices such as mobile stations, thus restricting the performance of the antenna; this is exacerbated at lower frequencies, for example below 1 GHz, where the size of an antenna tends to become larger due to the longer wavelength. The receiver capability can therefore be degraded leading to decreased link performance.
To mitigate the above effects, the LNA noise figure should be as good as possible. However, achieving good noise performance without using external matching components prior to the LNA and with adequate current consumption is a challenging task. Additionally, as well as the expensive and size consuming external components, the cost of the RFIC containing the LNA must also be considered. To keep the semiconductor die area of the RFIC small, the number of on-chip inductors should be kept at a minimum, because high quality inductors require significant die area and their size does not downscale along with reductions in the features widths of integrated circuits.
From the above it can be seen that there are a number of different design factors to be considered when designing an LNA, and that accommodating some or all of these factors simultaneously can prove difficult. There is therefore a need to enhance LNA design by providing improved ways of accommodating various design factors.